This application tests a theory-based HIV treatment adherence intervention delivered by cell phone to patients in urban and rural areas. Adherence to antiretroviral (ART) medications is necessary to achieve HIV suppression and non-adherence can lead to treatment resistant genetic variants of HIV. People living with HIV/AIDS often experience difficulty sustaining high-levels of treatment adherence. Most factors that interfere with adherence are unanticipated and occur between clinical visits, including depression, side effects, substance use, and lapses in pharmacy refill. We will conduct a randomized clinical trial to test a cell phone-delivered theory-based medication adherence counseling intervention. The intervention is grounded in Self-Regulation Model and utilizes unannounced pill counts to monitor adherence and guide corrective feedback within the counseling context. Using pill count adherence information for counseling allows providers to detect and correct patient non-adherence within a time frame that can head off viral resistance. An experimental factorial design will test the effects of counseling and a text message reminder system as well the interaction of counseling and text message components with 600 men and women receiving HIV treatment. Following screening, informed consent and baseline assessments participants will be randomized to one of four intervention conditions:(a) self-regulation counseling + regimen tailored text message reminders; (b) self-regulation counseling + control messages; (c) control counseling + regimen tailored text reminders; or (d) control counseling + control messages. The study will therefore use a 2 (self-regulation counseling) x 2 (text message reminders) full factorial deign with participants followed for 12 months post-intervention. The primary endpoints are medication adherence assessed by unannounced pill counts and HIV RNA (viral load). This newly developed intervention strategy is grounded in Self-Regulation Theory and is designed for use in clinical settings with limited resources.